Given the recent advances in identification, separation and culture of human dendritic cells, it is likely that clinical trials using adoptively transferred cells will play an important role during the next stages in the development of adoptive immunotherapy. The major problem associated with current adoptive therapies (which have been limited to predominantly patients with melanoma and renal cell carcinoma) is that they utilize T- cells which have been expanded from tumor sites where they are admittedly ineffective. Developing new T-cell (and potentially B-cell) reactivity to tumor is a major goal of DC-based strategies. We have designed a clinical protocol in which HLA-A2+ patients with melanoma will have peripheral blood derived dendritic cells cultured with GM-CSF and IL-4 for 5-7 days, pulsed with synthetic melanoma peptides derived from the recently identified MART1/Melan A, gp100, or tyrosinase proteins. More problematic, and the major focus of this application, has been the derivation of tumor antigens from tumors in which relevant T-cell epitopes have not been identified and for the HLA-A2 negative individuals with melanoma. Unanswered questions are those related to the number of tumor cells required, the optimum strategy for delivery (timing, route, method, and type of natural antigen-following in vitro manipulation with cytokines or heat, or route, method, and type of natural antigen-following in vitro manipulation with cytokines or heat, or subsequent to genetic modification), and the appropriate cytokine stimulation. Furthermore, information about the status of dendritic cells in the peripheral blood of cancer patients, the possibility of dendritic cell dysfunction related to factors at the tumor site, and the optimum site, and the optimum means to expand cells in vivo for clinical dysfunction related to factors at the tumor site, and the optimum means to expand cells in vivo for clinical trials (GM-CSF, IL-4, Flt-3L) need to be clarified and defined. We will perform a series of studies designed to investigate these issues in patients with cancer. We will develop the following specific aims: AIM I. Determine to investigate these issues in patients with cancer. We will develop effective means to derive and deliver tumor antigen peptide or proteins to autologous dendritic cells to elicit tumor-specific T-cells in cancer patients. AIM II. Use genetic modifications to enhance the function of adoptively transferred DC's by the expression of cytokines or tumor antigens.